original article novel and reversible mechanisms of smoking ... - diabetes€¦ · smoking is the...

Novel and Reversible Mechanisms of Smoking-InducedInsulin Resistance in HumansBryan C. Bergman, Leigh Perreault, Devon Hunerdosse, Anna Kerege, Mary Playdon, Ali M. Samek,

and Robert H. Eckel

Smoking is the most common cause of preventable morbidity andmortality in the United States, in part because it is an independentrisk factor for the development of insulin resistance and type 2diabetes. However, mechanisms responsible for smoking-inducedinsulin resistance are unclear. In this study, we found smokers wereless insulin sensitive compared with controls, which increased aftereither 1 or 2 weeks of smoking cessation. Improvements in insulinsensitivity after smoking cessation occurred with normalization ofIRS-1ser636 phosphorylation. In muscle cell culture, nicotine exposuresignificantly increased IRS-1ser636 phosphorylation and decreasedinsulin sensitivity, recapitulating the phenotype of smoking-inducedinsulin resistance in humans. The two pathways known to stim-ulate IRS-1ser636 phosphorylation (p44/42 mitogen-activated pro-tein kinase [MAPK] and mammalian target of rapamycin [mTOR])were both stimulated by nicotine in culture. Inhibition of mTOR, butnot p44/42 MAPK, during nicotine exposure prevented IRS-1ser636

phosphorylation and normalized insulin sensitivity. These dataindicate nicotine induces insulin resistance in skeletal muscleby activating mTOR. Therapeutic agents designed to oppose skeletalmuscle mTOR activation may prevent insulin resistance in humanswho are unable to stop smoking or are chronically exposed to sec-ondhand smoke. Diabetes 61:3156–3166, 2012

Although the prevalence of cigarette smoking hasfallen dramatically since the mid-1950s, it is stillwidespread in the United States, with ;20%prevalence reported by the Center for Disease

Control (1). Smoking is the most common cause of pre-ventable morbidity and mortality in the United States (2),and is an independent risk factor for the development ofcardiovascular disease (3) and type 2 diabetes (4,5). Car-rying an almost equal risk as active smoking for the de-velopment of cardiovascular disease is exposure tosecondhand smoke (6). Data from National Health andNutrition Examination Survey III for 1988 indicated that88% of the population had detectable serum cotinine (7),a by-product of nicotine metabolism, indicating an almostnationwide exposure to cigarette smoke and the increasedisease burden it brings. In 2002, still as much as 43% ofthe U.S. population had serum cotinine concentrations in-dicating secondhand smoke exposure (8). Therefore, de-spite efforts to decrease smoking and related diseases in theUnited States, we are still a nation with tobacco-relatedmorbidity and mortality affecting almost half of the pop-ulation. These epidemiologic observations underscore the

importance of understanding how smoking impacts cardio-vascular disease and diabetes risk. Unfortunately, the mo-lecular link among smoking, secondhand smoke, and insulinresistance is unknown.

Nicotine, one of many components of tobacco smoke, isassociated with decreased insulin sensitivity in humans andtherefore may link smoking with insulin resistance (9,10).Nicotine binds to nicotinic acetylcholine a1 receptors inhuman skeletal muscle and muscle cell cultures (11,12).However, molecular mechanisms explaining nicotine- andsmoking-induced insulin resistance (13–15) and the im-provement after smoking cessation (15,16) are still unclear.Previous data from our laboratory found young chronicsmokers were insulin-resistant compared with nonsmokers(17) with increased basal inhibition of insulin signaling andsaturation of intramuscular lipids. Therefore, we combineda smoking-cessation intervention with nicotine exposure inmuscle cell culture to elucidate mechanisms explainingsmoking-induced insulin resistance. Our efforts focused ondetermining how nicotine increases basal inhibition ofIRS-1 to lead the way for therapeutic strategies to increaseinsulin sensitivity in individuals who cannot stop smokingor are habitually exposed to secondhand smoke.

RESEARCH DESIGN AND METHODS

Subjects. Twelve healthy sedentary nonsmokers and 10 smokers completedthis intervention study (Table 1). Two subjects had urinary cotinine in-compatible with smoking cessation (.30 ng/mL) and were therefore excludedfrom the analysis, leaving eight smokers who completed the intervention (7).Subjects gave informed consent and were excluded if they: had diabetes, hy-perlipidemia, liver, kidney, thyroid, or lung disease, a BMI,20 or.25 kg/m2, orwere taking medications that can affect glucose or lipid metabolism. All sub-jects were sedentary and engaged in moderate to vigorous exercise ,3 h/week.Subjects were weight stable in the 6 months prior to participation in this re-search study. This study was approved by the Human Resources Committee atthe University of Colorado at Boulder and the Colorado Multiple InstitutionReview Board at the University of Colorado at Anschutz Medical Campus.Preliminary testing. Subjects reported to the General Clinical ResearchCenter (GCRC) for screening procedures following a 12-h overnight fast, wherethey were given a health and physical exam followed by a fasting blood draw.Body composition was determined using dual energy X-ray absorptiometryanalysis (Lunar DPX-IQ; Lunar Corporation, Madison, WI).Diet and exercise control.All subjects were given a prescribed diet for 3 daysprior to both metabolic studies on the GCRC, with energy intake and com-position as previously described (17). Subjects were asked to refrain fromplanned physical activity for 48 h before each metabolic study.Baseline metabolic study. Subjects arrived to the GCRC after an overnightfast at 7 A.M., when an antecubital vein in one arm was cannulated for isotopeinfusion, and a retrograde dorsal hand vein in the contralateral side wascatheterized for blood sampling via the heated hand technique. To measureglucose turnover and intramuscular triglyceride (IMTG) fractional synthesisrate (FSR), background blood and breath samples were taken, followed bya primed (4.5 mg/kg) constant (0.03 mg/kg/min) infusion of [6,6-2H2]glucose,and a continuous infusion of [U-13C] palmitate (Isotec, Miamisburg, OH)(0.0174 mmol/kg/min). These infusions were continued for 4 h and throughoutthe muscle biopsy procedure. Subjects remained fasting during the tracer in-fusion. During the 4-h infusion, smokers smoked one cigarette (Camel Light;R.J. Reynolds, Winston-Salem, NC) every 30 min until the start of the final blood

From the Department of Endocrinology, Diabetes, and Metabolism, Universityof Colorado Denver, Denver, Colorado.

Corresponding author: Bryan C. Bergman, [email protected] 3 April 2012 and accepted 3 July 2012.DOI: 10.2337/db12-0418� 2012 by the American Diabetes Association. Readers may use this article as

long as the work is properly cited, the use is educational and not for profit,and the work is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

See accompanying commentary, p. 3078.

3156 DIABETES, VOL. 61, DECEMBER 2012 diabetes.diabetesjournals.org

ORIGINAL ARTICLE

mailto:[email protected]

http://creativecommons.org/licenses/by-nc-nd/3.0/

http://creativecommons.org/licenses/by-nc-nd/3.0/

sampling for a total of eight cigarettes. Blood sampling was performed duringminutes 210, 220, 230, and 240 of infusion. A percutaneous needle biopsy wasperformed after 240 min of isotope infusion for determination of IMTG anddiacylglycerol (DAG) concentration, composition, and IMTG FSR (18). Musclebiopsies (;150 mg) were taken from midway between the greater trochanter ofthe femur and the patella from similar depth to minimize variance in musclefiber composition, which varies with depth and length in the vastus lateralis(19). Muscle was immediately flash frozen in liquid nitrogen and stored at280°C until dissection and analysis.

After the tracer infusion, insulin sensitivity was determined via an insulin-modified frequently sampled intravenous glucose tolerance test using standardmethods (20), as previously described (17).Smoking-cessation intervention. Ten days before the first tracer infusionstudy, smokers began taking buproprion (Zyban [GlaxoSmithKline]; 150 mgtwice daily) to increase the compliance and success rate with smoking ces-sation. Smokers were taking buproprion during the metabolic studies bothpre- and postcessation. Smokers stopped smoking for 1 (n = 4) or 2 (n = 4)weeks starting the day of the baseline metabolic study. The intervention wasbased on the American Cancer Society model and did not include the use ofnicotine gum or patches. Subjects stopping smoking reported to the GCRCevery 2 to 3 days to measure breath carbon monoxide (CO) and urine cotinineconcentration to document smoking cessation. During the intervention, con-trol subjects maintained their normal lifestyle.Follow-upmetabolic study. Subjects returned to the GCRC after 1 or 2 weeksof smoking cessation or normal living to repeat the metabolic study. Duringthis follow-up study, smokers who stopped smoking did not smoke ciga-rettes during the study, unlike their first visit. All other procedures wereidentical between the two visits.Metabolite and hormone analyses. Standard enzymatic assays were used tomeasure glucose (Olympus AU400e Chemistry analyzer; Olympus America Inc.,Center Valley, PA), lactate (Sigma Kit #826; Sigma-Aldrich, St. Louis, MO), andfree fatty acids (FFA) (NEFA Kit; Wako, Richmond, VA). Insulin, glucagon, andadiponectin were measured using a radioimmunoassay (Diagnostic SystemsLaboratories, Inc., Webster, TX), and catecholamines were measured usinghigh-performance liquid chromatography. Plasma interleukin-6 and tumor ne-crosis factor-a (TNF-a) were measured using an enzyme-linked immunosorbentassay kit (R&D Systems HS600B and HSTA00C, respectively; R&D Systems,Minneapolis, MN). Breath CO was measured using an EC50-Micro Smokerlyzer(Bedfont Scientific, Medford, NJ). Urine cotinine was measured using an enzyme-linked immunosorbent assay (Calbiotech, Inc., Spring Valley, CA).Muscle lipid analysis. Skeletal muscle samples were dissected free ofextramuscular fat on ice, lyophilized, and processed as previously described toisolate intramuscular FFA, IMTG, and DAG concentration, enrichment, andcomposition (17).

Western blotting. Frozen skeletal muscle samples were prepared for Westernblot analysis and run on SDS-PAGE gels using standard techniques as describedpreviously (17).Plasma palmitate and glucose isotope analysis. Plasma glucose de-rivatization and analysis was determined as previously described (21). Meth-ylation and extraction of plasma palmitate was performed as previouslydescribed (22). Samples were run on an HP 6890 GC with a 30 m DB-23capillary column (Hewlett Packard, Palo Alto, CA) connected to an HP 5973MS (Hewlett Packard). Enrichments were calculated based on a standardcurve of known enrichments and corrected for variations in abundance (23).Peak identities were determined by retention time and mass spectra comparedwith standards of known composition.Isotope ratio mass spectrometry. To measure whole-body palmitate oxi-dation rates, 2 mL of breath CO2 was transferred into a 20-mL exetainer for themeasurement of 13CO2/

12CO2 with continuous flow isotope ratio mass spec-trometry (D V; Thermo Electron, Bremen, Germany). Each sample wasinjected (1.2 mL/injection) in duplicate for isotope ratio analyses, with an av-erage SD for all injections of 0.0001 atom percent.Cell culture. The rat myotube L6 cell line was purchased (American TypeCulture Collection, Manassas, VA) and grown to 60% confluence on six-wellplates using Dulbecco’s modified Eagle’s medium containing 10% FBS, 0.5%fungizone, 200 units/mL penicillin, and 200 mg/mL streptomycin at 37°C underand atmosphere of 5% CO2 and 95% air. Differentiation was initiated byswitching medium to skeletal muscle differentiation medium containing 2%horse serum, 0.5% fungizone, 200 units/mL penicillin, and 200 mg/mL strepto-mycin. Cells from passages 2–4 were used for experiments. Experiments wereperformed between 8 and 10 days of differentiation. Nicotine was added at200 mmol final concentration for all experiments, which is roughly two timeshigher than a high plasma concentration for an active smoker (24,25). Inhib-itors of p44/p42 mitogen-activated protein kinase (MAPK)/extracellular signal-related kinase (ERK) 1/2 (MEK; UO126) and mammalian target of rapamycin(mTOR; rapamycin) were each dissolved in dimethyl sulfoxide (DMSO) at10 mmol final concentration for all experiments (both from Cell SignalingTechnology, Inc., Danvers, MA).Time-course experiments. Cells were serum starved for 3.5 h and thenpreincubated with DMSO control and inhibitors for 30 min. After 0, 5, 10, 30, 60,and 120 min exposure to nicotine, cells were harvested in homogenizationbuffer previously described (17). After protein extraction and determination ofprotein concentration, cells were run on an SDS-PAGE Western blot usingstandard methods previously described (17). Antibodies against IRS-1ser636,IRS-1 total, phospho- and total p44/42 MAPK, phospho-p70s6k, and p70s6ktotal were purchased from Cell Signaling Technology.Insulin-stimulated glucose uptake. Cells were serum-starved for a total of4 h prior to determination of insulin sensitivity using 3H-2-deoxyglucose uptake

TABLE 1Subject demographics

Variable

Baseline Postintervention

Controls Smokers Controls Smokers

n (men/women) 12 (8/4) 8 (7/1)Age (years) 20.4 6 0.6 21.8 6 0.7Body fat (%) 19.5 6 2.2 18.5 6 1.9BMI (kg/m2) 21.3 6 0.5 21.9 6 0.9 21.1 6 0.6 21.8 6 1.0Body weight (kg) 64.2 6 2.8 69.7 6 2.8 64.7 6 3.6 69.4 6 2.6Fasting glucose (mg/dL) 84.3 6 2.5 86.6 6 0.9 80.7 6 1.6 87.8 6 1.0FFA (mmol/L) 657 6 81 657 6 51 639 6 69 591 6 65Lactate (mmol/L) 0.71 6 0.1 0.81 6 0.1 0.73 6 0.1 0.66 6 0.1Insulin (mU/mL) 4.7 6 0.6 5.5 6 0.5 4.9 6 0.5 5.3 6 0.7Glucagon (pg/mL) 56 6 3 51 6 3 59 6 3 54 6 4TGs (mg/dL) 59 6 8 116 6 9§ 65 6 9 135 6 13§Norepinephrine (pg/mL) 221 6 16 247 6 24 202 6 17 211 6 26Epinephrine (pg/mL) 28 6 2 27 6 4 23 6 2 28 6 3Adiponectin (pg/mL) 10.6 6 1 11.1 6 1 10.9 6 1 11.0 6 2TNF-a (pg/mL) 1.5 6 0.2 1.6 6 0.3 1.5 6 0.1 1.4 6 0.1Interleukin-6 (pg/mL) 1.8 6 0.4 3.6 6 1.3 2.5 6 0.4 2.2 6 0.5VO2 (L/min) 0.24 6 0.01 0.25 6 0.01 0.23 6 0.01 0.25 6 0.01Cigarettes/day 0 18.0 6 0.8§ 0 0Breath CO (ppm) 3.1 6 0.5 27.2 6 2.4§ 2.6 6 0.3 3.3 6 0.3*

Values are means 6 SEM. §Significantly different between nonsmokers and smokers, P , 0.05. *Significantly different after the intervention,P , 0.05.

B.C. BERGMAN AND ASSOCIATES

diabetes.diabetesjournals.org DIABETES, VOL. 61, DECEMBER 2012 3157

as previously described (26). DMSO vehicle and inhibitors were added after1.5 h of serum starving and nicotine or nicotine vehicle added after 2 h of serumstarving. Concentrations of vehicle, inhibitors, and nicotine were maintainedduring preincubation of cells with insulin and measurement of glucose uptake.Measures were performed in duplicate per assay.Basal IRS-1

ser636phosphorylation. Vehicle and inhibitors were added after

1.5 h of serum starving and nicotine or nicotine vehicle control added after 2 h ofserum starving. After 4 h of serum starving and treatments, cells were harvestedon ice, protein run on SDS-PAGEWestern blots, and probed as described above.Calculations. IMTG FSR and IMTG and DAG saturation were calculated aspreviously described (17). Palmitate and glucose rate of disappearance andpalmitate rate of oxidation were calculated using steady-state kinetics anda whole-body estimate of carbon label retention as previously described (27).Statistics. Data are presented as mean 6 SEM. Differences in normally dis-tributed data between smokers and nonsmokers, before and after the smoking-cessation intervention, were analyzed using a two-way repeated-measuresANOVA (JMP, Cary, NC). Nonnormally distributed data were log transformedprior to analysis. When the repeated-measures ANOVA indicated a significantinteraction, changes before and after smoking cessation were determinedusing a paired t test. Differences in muscle cell culture were determined usingunpaired t tests. An a level of 0.05 was used throughout.

RESULTS

All subjects were college-aged men and women with similarBMI, percent body fat, blood metabolites, and metabolic rate(Table 1). On average, smokers reported smoking almostone pack of cigarettes/day. The period of smoking cessationwas only 1 to 2 weeks in order to minimize changes in bodyweight that could have confounded our results. We weresuccessful, as there was no significant change in bodyweight during the intervention (Table 1). There were nodifferences in 1 to 2 weeks of smoking cessation, so datawere combined to gain power for planned analyses.Smoking-cessation intervention. Fasting blood triglyc-eride (TG) concentration was roughly twice as high in smokerscompared with nonsmokers (P , 0.0001), which did notchange after smoking cessation (Table 1). As expected,smokers had greater breath CO measured during the firstmetabolic study, which decreased significantly and wasnot different from nonsmokers after smoking cessation

(Fig. 1A). Urinary cotinine is a better measure of smokingcessation compliance due to a longer half-life. Baselinecotinine was significantly elevated compared with non-smokers and decreased progressively to values not sig-nificantly different than nonsmokers after 7 and 14 days ofsmoking cessation (Fig. 1B).Smoking cessation increases insulin sensitivity. In-sulin sensitivity was significantly lower in smokers com-pared with nonsmokers (Fig. 2, P = 0.03) and increasedsignificantly in smokers after smoking cessation (P = 0.006).Insulin sensitivity was not different in nonsmokers after theintervention (P = 0.24). There were no significant differencesin glucose effectiveness (the ability of glucose to promote itsown transport) or the disposition index (insulin secretionadjusted for insulin resistance) between smokers and non-smokers before or after the smoking cessation intervention.

Similar to metabolic rate, whole-body resting fat oxida-tion was not significantly different between groups (controlsubjects = 1.4 6 0.1, smokers = 1.3 6 0.1 mmol/kg/min; P =0.84) and did not change significantly after the intervention.Palmitate rate of appearance (Ra) was significantly higherin smokers compared with nonsmokers and remainedhigher after smoking cessation (Fig. 3A, P = 0.004). The rateof oxidation of plasma palmitate was not significantly dif-ferent in smokers compared with nonsmokers (Fig. 3B, P =0.23) and did not change after smoking cessation. GlucoseRa was also not different between smokers and controls atbaseline (Fig. 3C, P = 0.18) and did not change significantlyafter the intervention (P = 0.18).Muscle lipid metabolism after smoking cessation. Wedid not find a difference in the resting concentration ofIMTG or DAG (Fig. 4A and D) or FSR of IMTG (Fig. 4B)in smokers compared with nonsmokers before or aftersmoking cessation. However, the composition of IMTGand DAG differed between groups, with significantlygreater proportion of saturated acyl side chains in smokerscompared with nonsmokers before and after smokingcessation (Fig. 4C, P = 0.02, and Fig. 4E, P = 0.04).

FIG. 1. Time-course breath CO (A) and urinary cotinine (B) concentration at baseline and during smoking cessation. Values are means 6 SEM.

SMOKING CESSATION AND INSULIN ACTION


Alterations in protein content after smoking cessation.Western blot analyses were performed to measure poten-tial mechanisms promoting the increase in insulin actionafter smoking cessation. Table 2 shows a summary of theseanalyses that were not significantly different betweengroups or following smoking cessation. The only differencein protein expression was significantly greater content ofIRS-1ser636 phosphorylation in smokers compared with non-smokers (Fig. 5, P = 0.02), which decreased in smokers(P = 0.047) to values that were not significantly differentthan nonsmokers after the intervention.Nicotine-induced insulin resistance in muscle cellculture. Cell-culture experiments were performed in L6myotubes using nicotine exposure to determine mecha-nisms for smoking-induced IRS-1ser636 phosphorylation.We recapitulated human studies and found 2 h of nicotineexposure significantly increased IRS-1ser636 phosphoryla-tion (P = 0.03, Fig. 6A). Nicotine exposure only increasedIRS-1ser636 phosphorylation, whereas serine sites 307(P = 0.25), 612 (P = 0.29), and 1,101 (P = 0.72) were notsignificantly changed (Fig. 6A). Serine636 phosphorylationis promoted by p44/p42 MAPK and mTOR activation.

Therefore, we determined if nicotine stimulated thesepathways in cell culture. Nicotine exposure significantlyincreased phosphorylation of p44/p42 MAPK after 30 and60 min, but returned to baseline after 120 min (Fig. 6B).This effect could be blocked by the MEK inhibitor UO126.Nicotine exposure also increased mTOR activation from5–120 min, which could be blocked using the mTOR in-hibitor rapamycin (Fig. 6C).

Fig. 6D shows the nicotine-induced increase in IRS-1ser636

phosphorylation compared with control (P = 0.04), whichwas not altered by UO126 (P = 0.98) during nicotine ex-posure. However, rapamycin during nicotine exposure sig-nificantly decreased IRS-1ser636 phosphorylation comparedwith nicotine treatment alone (P = 0.03).

Nicotine administration recapitulated insulin resistancein L6 cells, as insulin-stimulated glucose uptake decreased57% with 2 h of nicotine exposure (Fig. 7). UO126 duringnicotine exposure did not improve insulin sensitivity.However, rapamycin administration during nicotine ex-posure significantly increased insulin sensitivity (P = 0.02)compared with nicotine alone to values not significantlydifferent than control levels.

FIG. 2. Insulin sensitivity measured using the Bergman minimal model in nonsmokers and smokers. Data for the change in each individualare shown. Values are means 6 SEM. §Significantly different than nonsmokers, P < 0.05; *significantly different after smoking cessation,P < 0.05. Si, insulin sensitivity index.



DISCUSSION

The detrimental effect of cigarette smoking on insulin sen-sitivity is well documented. Smoking-induced insulin re-sistance has been reported in epidemiologic (28) as well ascross-sectional studies in middle-aged people (13,14). We

previously reported insulin resistance in young, otherwisehealthy college-aged men and women through inhibition ofbasal insulin signaling in skeletal muscle (17). This studyreports on a cohort of subjects studied after 1 to 2 weeksof verified smoking cessation. Insulin sensitivity increasedsignificantly after smoking cessation. The concentration ofplasma TGs and saturation of intracellular lipids remainedhigher in ex-smokers compared with nonsmokers, butbasal inhibition of insulin signaling reverted to normal. Cell-culture methods recapitulated insulin resistance with nico-tine exposure and showed nicotine stimulated p44/p42MAPK and mTOR. Only inhibition of mTOR activation wasable to prevent nicotine-induced insulin resistance. There-fore, individuals who cannot stop smoking or who are ex-posed to second hand smoke may benefit from therapeuticagents designed to prevent mTOR activation to reduce theirrisk of type 2 diabetes.

The current data show 1 to 2 weeks of smoking cessa-tion, without changes in body weight or adiposity, resultedin increased insulin sensitivity in a young population.These data are similar to others showing increased insulinsensitivity after short (13) and long-term (16) smokingcessation in older individuals. However, these data con-trast reports that smoking cessation did not change glu-cose tolerance after 4 months (29) or increased risk of type2 diabetes over 3 years, both of which were largely drivenby weight gain (30). Insulin secretion was appropriate forthe insulin resistance before and after smoking cessation,as revealed by unchanged disposition index scores. Com-bined, these data indicate that smoking cessation, withouta change in body weight, may decrease diabetes risk.However, short-term smoking cessation did not com-pletely normalize insulin sensitivity, indicating persistentmetabolic alterations in ex-smokers.

Several metabolic variables remained significantly dif-ferent in ex-smokers compared with nonsmokers, includingincreased plasma TG concentration and palmitate turnover,which may reflect or result in residual insulin resistance.Plasma TG concentration is likely only a marker of insulinresistance, as it is not thought to directly affect insulinsensitivity (31). Increased palmitate Ra, without differencesin fasting catecholamines, insulin, or glucagon betweengroups, suggests adipocyte insulin resistance in smokersbefore and after smoking cessation. In steady state, palmi-tate appearance is matched by palmitate disappearanceinto peripheral tissues. Higher rates of FFA disappearanceand nonadipocyte re-esterification have been reported insmokers, consistent with increased hepatic FFA uptakeand VLDL-TG production, or increased storage of FFA asTG in muscle (15). Our data eliminate muscle as a site ofincreased FFA re-esterification before and after smokingcessation and point to increased hepatic FFA uptake andVLDL-TG production. These data suggest that future studiesinvestigating differences in hepatic lipid content in smokersversus nonsmokers may be warranted. Short-term smokingcessation did not decrease basal palmitate turnover, in-dicating prolonged smoking cessation is required to nor-malize adipocyte insulin resistance. Increased systemic FFArelease and VLDL-TG production may play an importantrole linking cigarette smoking and increased risk of car-diovascular disease (3).

In addition to increases in FFA turnover, cigarettesmoking has been reported to increase plasma FFA con-centration and skeletal muscle lipoprotein lipase withoutchanging whole-body fat oxidation (15). This metabolicmilieu could lead to alterations in intramuscular lipid

FIG. 3. Palmitate Ra (A), oxidation (B), and glucose Ra (C) before andafter smoking cessation intervention in control subjects and cigarettesmokers. Values are means6 SEM. §Significantly different than controlsubjects, P < 0.05.



metabolism promoting insulin resistance in smokers,which could normalize after smoking cessation. However,our data suggest alterations in IMTG and DAG concen-tration and IMTG synthesis rates are not responsible forthe increase in insulin sensitivity after short-term smok-ing cessation. Interestingly, saturation of IMTG and DAGremained significantly increased in smokers after smoking

cessation. Although the implications of increased IMTGsaturation are currently unknown, increased DAG sat-uration is related to insulin resistance even withouta change in total concentration (32–34), possibly byaltering protein kinase C activation (35,36). Increasedmuscle lipid saturation may also reflect increased satu-ration of other intracellular lipids, such as ceramide (37)

FIG. 4. IMTG concentration (A), FSR (B), and saturation (C) and intramuscular DAG concentration (D) and saturation (E) in nonsmokers andsmokers. Values are means 6 SEM. §Significantly different than nonsmokers, P < 0.05.



and long chain acyl-CoA (38), which may also promoteinsulin resistance. Therefore, persistently greater satura-tion of muscle lipids after smoking cessation may helpexplain why insulin sensitivity did not completely nor-malize in ex-smokers.

No differences were observed between smokers andnonsmokers, and no changes were found following smokingcessation in protein abundance associated with inflam-mation, lipid peroxidation, and endoplasmic reticulum stress.As previously reported (17), smokers had greater IRS-1ser636

TABLE 2Summary of Western blot data from human studies

Variable

Baseline Postintervention

Controls Smokers Controls Smokers

GRP78 0.94 6 0.12 0.84 6 0.13 0.93 6 0.11 1.06 6 0.1MAP4K4 1.16 6 0.20 1.41 6 0.38 1.36 6 0.12 1.51 6 0.234-HNE 1.05 6 0.17 0.97 6 0.15 1.24 6 0.16 1.11 6 0.07SDH 1.25 6 0.08 1.13 6 0.10 1.25 6 0.08 1.13 6 0.08PGC1a 0.6 6 0.05 0.68 6 0.08 0.67 6 0.06 0.76 6 0.06TLR2 0.95 6 0.11 1.1 6 0.16 1.09 6 0.1 1.2 6 0.2Phospho-AMPK/total 1.13 6 0.23 0.84 6 0.24 1.0 6 0.14 1.05 6 0.30Phospho IKKa/total 1.43 6 0.37 1.32 6 0.50 1.47 6 0.28 1.36 6 0.37Phospho-ser307/IRS1 total 0.84 6 0.22 1.1 6 0.22 0.74 6 0.12 0.74 6 0.11Phospho-p44 MAPK/total 0.46 6 0.07 0.47 6 0.18 0.49 6 0.10 0.62 6 0.17Phospho-JNK/total 0.65 6 0.05 0.70 6 0.08 0.67 6 0.05 0.70 6 0.10Phospho-p70S6k/total 0.80 6 0.34 0.76 6 0.11 0.98 6 0.10 0.89 6 0.10

Values are means of arbitrary units 6 SEM. AMPK, AMP-activated protein kinase; IKKa, inhibitor of kB kinase a; JNK, Jun NH2-terminalkinase.

FIG. 5. IRS-1 serine636

phosphorylation before and after smoking cessation in control subjects and cigarette smokers. Values are means 6 SEM.§Significantly different than control subjects, P < 0.05; *significantly different after smoking cessation, P < 0.05.



phosphorylation compared with nonsmokers when theywere actively smoking. Serine phosphorylation of manysites on IRS-1, including ser636, decreases IRS-1 tyrosinephosphorylation, phosphatidylinositol-3-kinase binding,and insulin-stimulated glucose uptake (39). Following smokingcessation, IRS-1ser636 phosphorylation decreased in smokerssuch that there were no differences between groups afterthe intervention. Basal inhibition of IRS-1 by phosphoryla-tion of ser636 has been reported in other insulin-resistantstates such as type 2 diabetes (40) and in insulin-resistantoffspring of patients with type 2 diabetes (41). These datasuggest that basal inhibition of insulin signaling may be animportant mechanism responsible for smoking-induced in-sulin resistance and could also play a role in insulin resis-tance of secondhand smoke exposure.

Several pathways promote increased IRS-1ser636 phosphor-ylation that may be implicated in the insulin resistance ofsmoking. This specific serine site on IRS-1 can be phosphor-ylated by p44/p42 MAPK (40) and mTOR (42). Neither of thesepathways was significantly altered in biopsies taken fromsmokers 45–60 min after smoking their last cigarette or aftersmoking cessation. These data suggest smoking may resultin transitory stimulation of these pathways, which results indownstream serine phosphorylation of IRS-1 despite a returnof the pathway initiating the change to control levels.

Based on this idea of transitory stimulation of insulindesensitizing pathways from smoking, we sought to evaluatemechanisms by which nicotine may promote IRS-1ser636

phosphorylation in L6 muscle culture. Nicotine was usedbecause the majority of smoking-induced insulin resistancecan be reproduced with transdermal nicotine administrationin humans and thus is thought to be an important componentof tobacco smoke promoting insulin resistance (10). How-ever, nicotine infusion or transdermal administration doesnot acutely change insulin sensitivity in healthy individuals(43,44), whereas chronic use of nicotine gum appears topromote insulin resistance (9). The acute effects of nicotinediverge in insulin-resistant individuals, in whom nicotine in-fusion or transdermal administration decreases insulin sen-sitivity (10,43). Therefore, the literature suggests nicotineadministration in nicotine naive individuals may requirechronic exposure to impact insulin sensitivity, whereas anacute exposure is enough to decrease insulin sensitivity inindividuals with underlying insulin resistance. In muscle cellculture, nicotine transiently stimulated p44/p42 MAPK andcould be prevented by inhibition with UO126. However,preventing activation of p44/p42 MAPK by nicotine did notincrease insulin sensitivity or alter IRS-1ser636 phosphoryla-tion. Therefore, nicotine-induced insulin resistance is notlikely to result from p44/p42 MAPK activation.

FIG. 6. Effects of nicotine on cell signaling and insulin sensitivity in L6 skeletal muscle myotubes. A: Serine phosphorylation of IRS-1 in response tonicotine exposure at 200 mmol. B: Stimulation of p44/p42 MAPK over time with continuous nicotine exposure at 200 mmol with and without MEKinhibitor UO126, the upstream kinase of p44/p42 activation. C: Stimulation of p70s6k over time with continuous nicotine exposure at 200 mmolwith and without the mTOR inhibitor rapamycin. D: Effects of nicotine at 200 mmol on IRS-1

ser636phosphorylation in L6 muscle cell culture with

and without inhibitors against p44/p42 MAPK and mTOR. Values are mean 6 SEM. §Significantly different than control, P < 0.05; ¥significantlydifferent than time 0, P < 0.05; #significantly different than nicotine, P < 0.05.



Our cell-culture approach revealed mTOR as the mostlikely candidate to explain increased basal inhibition ofinsulin signaling in smokers. This idea is supported bycancer literature showing mTOR activation in lung lesionsfrom heavy smokers and clinical trials evaluating mTORinhibitors for the treatment of tobacco-induced lungtumorigenesis (45). Activation of mTOR in humans hasalso been implicated in insulin resistance of nutritional

abundance (46), which can be prevented with rapamycin(47). Similarly, in culture, rapamycin prevented nicotine-induced mTOR activation, prevented nicotine-induced IRS-1ser636 phosphorylation, and normalized insulin sensitivityduring nicotine exposure. Therefore, mTOR plays an im-portant role in the insulin resistance of smoking. These dataagree with a recent cell-culture study reporting nicotineincreased TNF-a expression through increased oxidativestress in the presence of palmitate, which lead to insulinresistance (48). Interestingly, TNF-a activates mTOR (42) andmay be one mechanism explaining reversible IRS-1ser636

phosphorylation and insulin resistance. We did not findchanges in plasma TNF-a after smoking cessation in thecurrent study, suggesting nicotine may promote local in-creases in muscle TNF-a. In individuals who smoke or areexposed to secondhand smoke, preventing nicotine-inducedmTOR activation in skeletal muscle could be a novel targetto decrease diabetes risk.

There are several important limitations with this study.The nicotine dose used in these studies of 200 mmol waschosen in order to maximize our ability to see the influenceof nicotine on insulin resistance. This concentration is ap-proximately double the highest dose found in the plasmaof smokers (24,25). Smoking-induced vascular dysfunctionhas been reported in some (49), but not all studies (50),and therefore may have influenced insulin sensitivity aftersmoking cessation. Although not measured in this study,smoking may increase sympathetic outflow in humans (51),which could reduce muscle blood flow and insulin sensi-tivity (52). Buproprion is not known to alter insulin sensi-tivity or muscle lipid metabolism, but theoretically couldhave influenced the results in smokers before or after

FIG. 7. Effects of nicotine, with and without inhibitors against p44/p42MAPK and mTOR, on insulin-stimulated glucose uptake in L6 musclecell culture. Values are mean 6 SEM. §Significantly different thancontrol, P < 0.05; #significantly different than nicotine, P < 0.05.

FIG. 6. Continued.



smoking cessation. Cell culture showed persistent activationof mTOR with continuous nicotine exposure, which doesnot agree with muscle biopsy data showing no change inphosphorylation of p70s6k, the downstream kinase formTOR, compared with controls or after smoking cessation.Therefore, the increase in mTOR activity from smoking maybe transitory in humans as compared with cell culture anddecrease to normal values 45–60 min after smoking a ciga-rette. Differences in nicotine clearance may change resultsof studies in culture compared with in vivo. Nicotine ismetabolized by the liver with a half-life of 2 h (53), thereforewithout hepatic clearance conditions in culture will notcompletely mimic the in vivo condition. Our cell-culturework used an acute administration of a constant dose ofnicotine on cell signaling. Chronic exposure, in addition tocyclical increases and decreases in nicotine from cigarettesmoking or secondhand smoke exposure, may elicit differ-ent responses.

In conclusion, chronic cigarette smokers had lower in-sulin sensitivity compared with nonsmokers, which im-proved but did not normalize after 1 to 2 weeks of smokingcessation. Greater saturation of skeletal muscle lipids wasmaintained after smoking cessation and may partially ex-plain the residual insulin resistance observed. IncreasedIRS-1ser636 phosphorylation in smokers normalized aftersmoking cessation, which may explain the increase in in-sulin sensitivity. Nicotine in muscle cell culture recapitulatedsmoking-induced insulin resistance and stimulated majorpathways promoting IRS-1ser636 phosphorylation (p44/p42MAPK and mTOR). However, only preventing mTOR acti-vation mitigated IRS-1ser636 phosphorylation and nicotine-induced insulin resistance in culture. Therapeutic agentsdesigned to oppose skeletal muscle mTOR activation mayprove a novel strategy for individuals who are at increasedrisk of diabetes and cardiovascular disease because theycannot stop smoking or are exposed to secondhand smoke.

ACKNOWLEDGMENTS

This work was partially supported by the National Insti-tutes of Health General Clinical Research Center GrantRR-00036, National Institute of Diabetes and Digestive andKidney Diseases grants (DK-064811 to L.P., DK-26356 toR.H.E., and DK-059739 to B.C.B.), and a Colorado TobaccoResearch Program grant (3K-027 to B.C.B.).

No potential conflicts of interest relevant to this articlewere reported.

B.C.B. designed the study, performed subject testing,performed cell culture experiments, analyzed data, andwrote the manuscript. L.P. helped design the study, pro-vided medical oversight, performed all biopsies, and helpedwrite the manuscript. D.H. and M.P. performed subjecttesting, analyzed samples, and edited the manuscript. A.K.performed subject testing, analyzed samples, helped withcell culture experiments, and edited the manuscript. A.M.S.performed subject testing and edited the manuscript. R.H.E.helped design the study and edited the manuscript. B.C.B. isthe guarantor of this work and, as such, had full access toall the data in the study and takes responsibility for theintegrity of the data and the accuracy of the data analysis.

Data from this study were presented in abstract form atthe 71st Scientific Sessions of the American DiabetesAssociation, San Diego, California, 24–28 June 2011.

The authors thank Dr. Dennis Peterson at the Universityof Colorado Health Sciences Center for the generousdonation of the 4-HNE antibody.

REFERENCES

1. Centers for Disease Control and Prevention (CDC). Cigarette smoking amongadults—United States, 2007. MMWRMorb Mortal Wkly Rep 2008;57:1221–1226

2. Schroeder SA, Warner KE. Don’t forget tobacco. N Engl J Med 2010;363:201–204

3. Kannel WB, D’Agostino RB, Belanger AJ. Fibrinogen, cigarette smoking,and risk of cardiovascular disease: insights from the Framingham Study.Am Heart J 1987;113:1006–1010

4. Feskens EJ, Kromhout D. Cardiovascular risk factors and the 25-year in-cidence of diabetes mellitus in middle-aged men. The Zutphen Study. Am JEpidemiol 1989;130:1101–1108

5. Perry IJ, Wannamethee SG, Walker MK, Thomson AG, Whincup PH, ShaperAG. Prospective study of risk factors for development of non-insulin de-pendent diabetes in middle aged British men. BMJ 1995;310:560–564

6. Barnoya J, Glantz SA. Cardiovascular effects of secondhand smoke: nearlyas large as smoking. Circulation 2005;111:2684–2698

7. Pirkle JL, Flegal KM, Bernert JT, Brody DJ, Etzel RA, Maurer KR. Expo-sure of the US population to environmental tobacco smoke: the ThirdNational Health and Nutrition Examination Survey, 1988 to 1991. JAMA1996;275:1233–1240

8. Pirkle JL, Bernert JT, Caudill SP, Sosnoff CS, Pechacek TF. Trends in theexposure of nonsmokers in the U.S. population to secondhand smoke:1988-2002. Environ Health Perspect 2006;114:853–858

9. Eliasson B, Taskinen MR, Smith U. Long-term use of nicotine gum is associ-ated with hyperinsulinemia and insulin resistance. Circulation 1996;94:878–881

10. Epifano L, Di Vincenzo A, Fanelli C, et al. Effect of cigarette smoking andof a transdermal nicotine delivery system on glucoregulation in type 2diabetes mellitus. Eur J Clin Pharmacol 1992;43:257–263

11. Albuquerque EX, Pereira EF, Alkondon M, Rogers SW. Mammalian nico-tinic acetylcholine receptors: from structure to function. Physiol Rev 2009;89:73–120

12. Keli�c S, Olsson T, Kristensson K. Interferon-gamma promotes proliferationof rat skeletal muscle cells in vitro and alters their AChR distribution. JNeurol Sci 1993;114:62–67

13. Attvall S, Fowelin J, Lager I, Von Schenck H, Smith U. Smoking inducesinsulin resistance—a potential link with the insulin resistance syndrome. JIntern Med 1993;233:327–332

14. Eliasson B, Attvall S, Taskinen MR, Smith U. The insulin resistance syn-drome in smokers is related to smoking habits. Arterioscler Thromb 1994;14:1946–1950

15. Hellerstein MK, Benowitz NL, Neese RA, et al. Effects of cigarette smokingand its cessation on lipid metabolism and energy expenditure in heavysmokers. J Clin Invest 1994;93:265–272

16. Eliasson B, Attvall S, Taskinen MR, Smith U. Smoking cessation improvesinsulin sensitivity in healthy middle-aged men. Eur J Clin Invest 1997;27:450–456

17. Bergman BC, Perreault L, Hunerdosse DM, Koehler MC, Samek AM, EckelRH. Intramuscular lipid metabolism in the insulin resistance of smoking.Diabetes 2009;58:2220–2227

18. Bergström J, Hermansen L, Hultman E, Saltin B. Diet, muscle glycogen andphysical performance. Acta Physiol Scand 1967;71:140–150

19. Lexell J, Henriksson-Larsén K, Winblad B, Sjöström M. Distribution ofdifferent fiber types in human skeletal muscles: effects of aging studied inwhole muscle cross sections. Muscle Nerve 1983;6:588–595

20. Boston RC, Stefanovski D, Moate PJ, Sumner AE, Watanabe RM, BergmanRN. MINMOD Millennium: a computer program to calculate glucose ef-fectiveness and insulin sensitivity from the frequently sampled intravenousglucose tolerance test. Diabetes Technol Ther 2003;5:1003–1015

21. Bergman BC, Cornier MA, Horton TJ, Bessesen DH. Effects of fasting oninsulin action and glucose kinetics in lean and obese men and women. AmJ Physiol Endocrinol Metab 2007;293:E1103–E1111

22. Patterson BW, Zhao G, Elias N, Hachey DL, Klein S. Validation of a newprocedure to determine plasma fatty acid concentration and isotopic en-richment. J Lipid Res 1999;40:2118–2124

23. Patterson BW, Zhao G, Klein S. Improved accuracy and precision of gaschromatography/mass spectrometry measurements for metabolic tracers.Metabolism 1998;47:706–712

24. Kilaru S, Frangos SG, Chen AH, et al. Nicotine: a review of its role inatherosclerosis. J Am Coll Surg 2001;193:538–546

25. Schievelbein H, Eberhardt R, Löschenkohl K, Rahlfs V, Bedall FK. Ab-sorption of nicotine through the oral mucosa. I. Measurement of nicotineconcentration in the blood after application of nicotine and total particu-late matter. Agents Actions 1973;3:254–258

26. Klip A, Gumà A, Ramlal T, Bilan PJ, Lam L, Leiter LA. Stimulation ofhexose transport by metformin in L6 muscle cells in culture. Endocrinol-ogy 1992;130:2535–2544



27. Wolfe R. Radioactive and Stable Isotope Tracers in Biomedicine: Prin-ciples and Practice of Kinetic Analysis. New York, Wiley-Liss, 1992

28. Manson JE, Ajani UA, Liu S, Nathan DM, Hennekens CH. A prospectivestudy of cigarette smoking and the incidence of diabetes mellitus amongUS male physicians. Am J Med 2000;109:538–542

29. Nilsson P, Lundgren H, Söderström M, Fagerström KO, Nilsson-Ehle P.Effects of smoking cessation on insulin and cardiovascular risk factors—a controlled study of 4 months’ duration. J Intern Med 1996;240:189–194

30. Yeh HC, Duncan BB, Schmidt MI, Wang NY, Brancati FL. Smoking,smoking cessation, and risk for type 2 diabetes mellitus: a cohort study.Ann Intern Med 2010;152:10–17

31. Shulman GI. Cellular mechanisms of insulin resistance. J Clin Invest 2000;106:171–176

32. Bergman BC, Perreault L, Hunerdosse DM, Koehler MC, Samek AM, EckelRH. Increased intramuscular lipid synthesis and low saturation relate to insulinsensitivity in endurance-trained athletes. J Appl Physiol 2010;108:1134–1141

33. van Hees AM, Jans A, Hul GB, Roche HM, Saris WH, Blaak EE. Skeletalmuscle fatty acid handling in insulin resistant men. Obesity (Silver Spring)2011;19:1350–1359

34. Bergman BC, Hunerdosse DM, Kerege A, Playdon MC, Perreault L. Lo-calisation and composition of skeletal muscle diacylglycerol predicts in-sulin resistance in humans. Diabetologia 2012;55:1140–1150

35. Hinderliter AK, Dibble AR, Biltonen RL, Sando JJ. Activation of proteinkinase C by coexisting diacylglycerol-enriched and diacylglycerol-poorlipid domains. Biochemistry 1997;36:6141–6148

36. Montell E, Turini M, Marotta M, et al. DAG accumulation from saturatedfatty acids desensitizes insulin stimulation of glucose uptake in musclecells. Am J Physiol Endocrinol Metab 2001;280:E229–E237

37. Holland WL, Brozinick JT, Wang LP, et al. Inhibition of ceramide synthesisameliorates glucocorticoid-, saturated-fat-, and obesity-induced insulinresistance. Cell Metab 2007;5:167–179

38. Houmard JA, Tanner CJ, Yu C, et al. Effect of weight loss on insulin sen-sitivity and intramuscular long-chain fatty acyl-CoAs in morbidly obesesubjects. Diabetes 2002;51:2959–2963

39. Gual P, Le Marchand-Brustel Y, Tanti JF. Positive and negative regulation ofinsulin signaling through IRS-1 phosphorylation. Biochimie 2005;87:99–109

40. Bouzakri K, Roques M, Gual P, et al. Reduced activation of phosphatidylinositol-3 kinase and increased serine 636 phosphorylation of insulin receptorsubstrate-1 in primary culture of skeletal muscle cells from patients withtype 2 diabetes. Diabetes 2003;52:1319–1325

41. Morino K, Petersen KF, Dufour S, et al. Reduced mitochondrial density andincreased IRS-1 serine phosphorylation in muscle of insulin-resistant off-spring of type 2 diabetic parents. J Clin Invest 2005;115:3587–3593

42. Ozes ON, Akca H, Mayo LD, et al. A phosphatidylinositol 3-kinase/Akt/mTOR pathway mediates and PTEN antagonizes tumor necrosis factorinhibition of insulin signaling through insulin receptor substrate-1. ProcNatl Acad Sci USA 2001;98:4640–4645

43. Axelsson T, Jansson PA, Smith U, Eliasson B. Nicotine infusion acutelyimpairs insulin sensitivity in type 2 diabetic patients but not in healthysubjects. J Intern Med 2001;249:539–544

44. Mora-Martínez JM, González-Ortiz M, Balcázar-Muñoz BR, Martínez-Abundis E. Acute effect of the transdermal administration of nicotine oninsulin sensitivity in healthy individuals with and without a family historyof type 2 diabetes mellitus in the first branch. Metab Syndr Relat Disord2004;2:227–233

45. Memmott RM, Dennis PA. The role of the Akt/mTOR pathway in tobaccocarcinogen-induced lung tumorigenesis. Clin Cancer Res 2010;16:4–10

46. Newgard CB, An J, Bain JR, et al. A branched-chain amino acid-relatedmetabolic signature that differentiates obese and lean humans and con-tributes to insulin resistance. Cell Metab 2009;9:311–326

47. Krebs M, Brunmair B, Brehm A, et al. The Mammalian target of rapamycinpathway regulates nutrient-sensitive glucose uptake in man. Diabetes2007;56:1600–1607

48. Tatebe J, Morita T. Enhancement of TNF-a expression and inhibition ofglucose uptake by nicotine in the presence of a free fatty acid in C2C12skeletal myocytes. Horm Metab Res 2011;43:11–16

49. Ijzerman RG, Serne EH, van Weissenbruch MM, de Jongh RT, StehouwerCD. Cigarette smoking is associated with an acute impairment of micro-vascular function in humans. Clin Sci (Lond) 2003;104:247–252

50. Rönnemaa EM, Rönnemaa T, Utriainen T, et al. Decreased blood flow butunaltered insulin sensitivity of glucose uptake in skeletal muscle ofchronic smokers. Metabolism 1999;48:239–244

51. Narkiewicz K, van de Borne PJ, Hausberg M, et al. Cigarette smoking in-creases sympathetic outflow in humans. Circulation 1998;98:528–534

52. Jamerson KA, Julius S, Gudbrandsson T, Andersson O, Brant DO. Reflexsympathetic activation induces acute insulin resistance in the humanforearm. Hypertension 1993;21:618–623

53. Benowitz NL, Jacob P 3rd, Jones RT, Rosenberg J. Interindividual vari-ability in the metabolism and cardiovascular effects of nicotine in man. JPharmacol Exp Ther 1982;221:368–372



original article novel and reversible mechanisms of smoking ... - diabetes€¦ · smoking is the...

Documents